JP3811238B2 - Voice guidance device for vehicles using image information - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an audio guidance device for a vehicle using image information for specifying and guiding a branch point using a feature extracted based on image information.
[0002]
[Prior art]
Conventionally, a navigation device is known as one of devices mounted on a vehicle and has been widely spread. This navigation apparatus has a normal display, and is based on displaying a map including a present location display here. When the destination is set, the optimum route to the destination is searched based on the stored map information. Then, in traveling in which a route to the destination is set, the route is always displayed and route guidance is performed.
[0003]
In addition, while driving, the driver needs to watch the front and cannot watch the display. Therefore, guidance such as turning left and right is also performed by voice. That is, a guidance voice such as “XXm ahead xx intersection is to the left” is output from the speaker in front of the guidance intersection (branch point) that needs guidance, and the route is guided. As a result, the driver can obtain guidance from the ear while gazing at the front.
[0004]
Here, the name of an intersection is usually used to specify a branch point, but it is not always easy to understand. Therefore, Japanese Patent Application Laid-Open No. 4-1520 and Japanese Patent Application Laid-Open No. 4-324212 propose to use a feature such as a building name existing at an intersection for guidance. By using such a feature, the branch point can be explained to the driver in an easy-to-understand manner, and more suitable route guidance can be performed.
[0005]
[Problems to be solved by the invention]
However, the one using the above-described feature constructs a guide sentence based on the presence of the feature in the two-dimensional map data. Therefore, when the guidance is output due to road undulations or the like, there is a possibility that the feature used for the guidance cannot be seen. In addition, due to construction or the like, the feature may not be visible, or the feature may not be visible due to the presence of a truck or bus in front. In such a case, the guidance using the characteristic object becomes very difficult to understand.
[0006]
The present invention has been made to solve the above-described problems, and an object thereof is to provide a vehicular voice guidance device using image information that can be guided by using features that are actually visible. To do.
[0007]
[Means for Solving the Problems]
The present invention includes a map database that stores map information including position information of a branch point, and a current position detection unit that detects a position of the vehicle that is the current position of the vehicle, and a route to the destination is set. in the running were at the guidance branch point requiring guidance and a branch point on the route, a vehicle audio guidance device that outputs voice guidance route guidance at the guidance branch point, the current position detecting means Based on the detection result and the map information stored in the map database, the guidance branch point recognition means for recognizing that the vehicle has approached the guidance branch point, the in-vehicle camera for imaging the front, and the guide branch point before measuring a positional relationship between the feature extracting means for extracting a characteristic object used to identify the guidance branch point in voice guidance from the imaging results of the in-vehicle camera, extracted feature thereof and the vehicle in The positional relationship between the guidance branch point and the characteristic object obtained from both the positional relationship measuring means, the positional relationship between the guidance branch point and the own vehicle position, and the positional relationship between the feature object and the own vehicle position. And determining means for determining that the feature is a feature existing in the vicinity of the guidance branch point based on information on the feature determined to be present in the vicinity of the guidance branch point by the determination means . The voice guidance specifying the guidance branch point is included in the voice guidance of the route guidance at the guidance branch point and output.
[0008]
As described above, in the present invention, the characteristic object is extracted from the imaging result obtained by the in-vehicle camera. Therefore, only features that are actually visible are used for guidance, and it is possible to reliably prevent guidance during construction or using items that cannot be seen by other vehicles.
[0010]
Then, the determination means, by determining the positional relationship between the vehicle and the characteristics thereof, characterized that is present in the image can be confirmed or not present in the branch point to be guided, the wrong characteristic object It is possible to effectively prevent using and guiding.
[0011]
Further, the present invention is characterized in that the positional relationship measuring means measures a distance between the own vehicle detected by stereo ranging of a plurality of in-vehicle cameras and a characteristic object. The distance to the feature can be recognized by the stereo distance measurement, and it can be reliably determined whether or not the feature exists at the branch point to be guided.
[0012]
The present invention is characterized in that the feature is a traffic light. Traffic lights are installed at many intersections and are very useful for identifying branch points.
[0013]
In the invention, it is preferable that the traffic light extracts a specific color portion and then specifies and detects the shape of the specific color portion by edge detection. Thus, the traffic light can be easily recognized by specifying the color.
[0014]
Further, the present invention is characterized in that color information in a traffic light is added to the voice guidance. Guidance can be provided by using the color of the traffic lights at that time, and guidance that is very easy to understand can be provided.
[0015]
The present invention is characterized in that the feature is a destination guide sign. The destination guide sign includes information on the course, and can be used for easy-to-understand guidance.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments (hereinafter referred to as embodiments) of the invention will be described with reference to the drawings.
[0017]
FIG. 1 is a block diagram showing the overall configuration of the apparatus of the present embodiment. The navigation control unit 10 performs various data processing for route guidance. The navigation control unit 10 is connected to a host vehicle position detection unit 12, and the navigation control unit 10 always recognizes the host vehicle position (current position) based on information from here. The own vehicle position detection unit 12 uses, for example, GPS (Global Positioning System), and a D-GPS device that takes into account error information provided by FM multiplex broadcasting can also be used. Furthermore, it is also preferable to combine position information supplied from an optical beacon or radio beacon, correction by map matching, autonomous navigation for detecting a position from a traveling direction and a travel distance, and the like.
[0018]
A map database 14 is connected to the navigation control unit 10. The map database 14 stores nationwide 2D map information, but may also store 3D information. That is, it is also preferable to store three-dimensional information including the three-dimensional shape data of roads such as three-dimensional intersections and the shapes of buildings around intersections. In particular, it is preferable to store information for displaying the shape of the intersection in three dimensions at the guidance intersection.
[0019]
In addition, the voice control unit 16 is connected to the navigation control unit 10 via the I / O interface 10a. The voice synthesizing unit 16 synthesizes the voice of the guidance sentence based on the signal from the navigation control unit 10 and outputs it from a speaker (not shown). The input operation unit 18 is for inputting a destination.
[0020]
Furthermore, this apparatus has two cameras 20a and 20b as in-vehicle cameras. The cameras 20a and 20b are CCD cameras that photograph the front of the vehicle, and are arranged at a predetermined distance in the vehicle width direction. Image processing units 22a and 22b are connected to the cameras 20a and 20b, respectively, and image signals, which are imaging results obtained by the cameras 20a and 20b, are supplied to the image processing units 22a and 22b. The image processing units 22a and 22b obtain digital image data based on the image signal. The image processing units 22a and 22b are connected to a recognition unit 24. The recognition unit 24 extracts and recognizes feature objects based on image data. Then, the recognition result of the recognition unit 24 is supplied to the navigation control unit 10. Further, the recognition unit 24 is connected to the navigation control unit 10, and the recognition result of the feature object is also supplied to the positional relationship measurement unit 26. The positional relationship measuring unit 26 stereoly calculates the position in the image of the same feature that is separately recognized from the imaging results of the cameras 20a and 20b, and the distance from the installation position of the cameras 20a and 20b to the feature. Measure distance. The positional relationship measuring unit 26 is also connected to the navigation control unit 10, and data about the measured distance and direction to the feature is also supplied to the navigation control unit 10.
[0021]
Next, the operation of this apparatus will be described based on the flowchart of FIG. First, when route guidance is used, a destination is first set and the current location is detected (S11). This destination is set by inputting an address or the like. For the input operation unit 18, a touch panel or the like is used. The current location can be obtained by taking in the output of the vehicle position detection unit 12.
[0022]
In this way, when the destination is set, the navigation control unit 10 searches for the optimum route from the current location to the destination based on the map data stored in the map database (S12). This route search is performed by a normal method using the Dijkstra method or the like.
[0023]
Then, when the obtained route is set and traveling toward the destination is started, the vehicle position is first detected (S13). Next, it is determined whether or not the vehicle is approaching a branch point (a branching intersection where guidance is required) (S14). That is, the branch point is known in advance on the route set to the destination. Therefore, it is determined whether the current location is approaching this branch point. It is determined from this determination result whether or not the vehicle has approached the branch point. If it has not approached the branch point, the process returns to S13, the current location is updated, and the determination of whether the branch point has been approached is repeated.
[0024]
When approaching the branch point, the approach to the branch point is guided along with the branch direction (S15). For example, normal guidance such as “500m ahead intersection is left direction” is performed.
[0025]
Next, the recognition unit 24 recognizes the images obtained from the cameras 20a and 20b, detects the presence or absence of a feature based on the result, and determines whether or not the feature exists (S16). As the characteristic object, a traffic light or the like can be considered. In the case of a traffic light, blue, red, and yellow are separately extracted by a color filter, and an edge of a portion with high luminance is detected. And since the shape of an edge is (circle), it determines with a traffic signal. In addition, it can be determined which signal is attached depending on which filter detects the circle.
[0026]
Furthermore, the direction and distance of the branch point seen from the own vehicle can be specified from the own vehicle position detecting means 12 and the map database 14, and the range on the image near the branch point to be guided can be estimated. Therefore, by performing the image processing only on that portion, the processing amount can be reduced and the processing speed can be improved. Further, by providing the cameras 20a and 20b with a steering mechanism, the vicinity of the target branch point can be captured near the center of the image. Therefore, it is possible to zoom up to capture the target, improve the resolution, and make the detection of the feature easier.
[0027]
Further, at the time of pattern matching in the recognition unit 24, the positional relationship between the branch point and the own vehicle is known. Therefore, it is possible to perform more efficient image processing by predicting the appearance of a feature using this positional relationship and performing pattern matching using a template corresponding to this. A plurality of templates may be provided according to each direction and distance, but the shape of the template may be changed according to the positional relationship.
[0028]
If there is a feature in S16, the positional relationship measuring unit 26 measures the distance and direction to the feature based on the images from the two cameras 20a and 20b, and the measured distance and Based on the direction, it is determined whether or not the feature exists at the intersection (branch point) (S17). That is, the relationship between the feature and the branch point is obtained from the relationship between the current location and the branch point, and the relationship between the current location and the feature object, and it is determined whether the feature object is on the branch point. Here, some features have a constant size. For example, many traffic lights have the same size. In this case, since the distance can be measured according to the size on the image plane, the stereo distance measurement by the two cameras becomes unnecessary.
[0029]
In the determination result of S17, if the feature is on the corresponding branch point, guidance can be performed using the feature. Therefore, the detected feature is stored as a candidate (S18). On the other hand, if it is determined in S17 that the feature is not a feature on the corresponding branch point, it is determined whether it is a destination guide sign in front of the branch point (S19). If it is a destination guidance sign, guidance is provided using the destination guidance sign at that stage (S20). That is, the destination guide sign is usually in front of the intersection. Therefore, the display contents of the destination guidance sign are acquired from the image or the map database 14 and the guidance such as “I am in the direction of XX according to the destination guidance sign” is performed at the stage where the destination guidance sign is recognized.
[0030]
If the feature in S18 is stored, if guidance in S20 is performed, if there is no feature in S16, or if it is not a destination guide sign in S19, there is sufficient room for branching. It is determined whether there is any (S21). This determination is performed, for example, by determining whether there is a point of 100 m to the branch point. However, on roads with many lanes, it is necessary to change lanes before branching, and it is considered that the time required for branching differs from roads on one lane. Therefore, it is preferable to change the determination as to whether there is a sufficient margin according to these conditions. In addition, the determination as to whether there is a margin may be changed according to the traveling speed or other traveling environment.
[0031]
If there is room in the determination in S21, the process returns to S16 and repeats the feature object detection process. On the other hand, when approaching the branch point and it is determined in S21 that there is no room, the optimum feature is selected from the stored candidate features and guidance is provided using this. It is not necessary to narrow down the features used for guidance. For example, an expression such as “turn right at the next signal following the vehicle ahead” is also possible (S22). For example, a guidance such as “Turn right at the next green light” is given. On the other hand, if there is no feature in S16 and if it is not a destination guide sign in S19, the feature cannot be found after all. For this reason, guidance is performed at a distance based on the vehicle position information. For example, a guidance “50m ahead and turn right” is given.
[0032]
In this way, when the guidance output is completed, it is determined whether or not the route has deviated (S23). If the route has deviated, the process returns to S12 and the route search is repeated. On the other hand, if it has not deviated, it is determined whether the destination has been reached (S24). If the destination has been reached, the processing is terminated. If the destination has not been reached in S24, the process returns to S13.
[0033]
As described above, in this embodiment, the feature object is extracted from the images obtained by the cameras 20a and 20b. Therefore, only features that are actually visible are used for guidance, and it is possible to reliably prevent guidance during construction or using items that cannot be seen by other vehicles. Furthermore, by measuring the distance and direction to the feature, it can be confirmed whether the feature existing in the image exists at the branch point, and erroneous guidance can be effectively prevented.
[0034]
Moreover, in the above-mentioned example, the traffic signal was adopted as a feature. Traffic lights are installed at many intersections and are very useful for identifying branch points. In addition, when signals are arranged close to each other, a plurality of traffic lights may be recognized by image processing. In this case, based on the distance and direction measured by the positional relationship measuring unit 26, it is recognized that there is another intersection in front of the intersection to be guided, and guidance such as “turn right at the second green light” is given. It is also suitable to do. Further, when there is no feature, normal guidance using the distance is performed, so that no special problem occurs even when there is no feature.
[0035]
"Other configurations"
In the above example, a traffic signal and a destination guide sign are taken up as features used for guidance. However, the feature is not limited to this, and various features can be adopted. For example, lane branch information (arrows in the lane), various traffic signs, pedestrian bridges, vehicles ahead, etc. can be adopted. In particular, for Y-shaped branching, guidance by recognizing a lane is suitable. It is possible to provide guidance such as “Turn right in this lane”. In addition, about a lane position, it is good to also detect the detection result of the own vehicle position detection part 12 collectively.
[0036]
Here, a building, a signboard, etc. can be used as a characteristic object. However, feature recognition is basically performed by pattern matching. That is, a feature is recognized based on a comparison between a shape previously held as a template and a shape obtained by image processing. Therefore, it is preferable to limit the characteristic objects to some objects whose shapes are known in advance. Furthermore, although it demonstrated also in the example of the traffic light, there is also a characteristic in which a color is known beforehand. For example, the destination guide sign can be identified by extracting the blue color and detecting the edge of the blue color.
[0037]
According to the apparatus of the present invention, sufficient route guidance can be performed basically only by voice guidance. Therefore, a display for route display can be omitted. As a result, it is possible to reduce the cost without taking up space.
[0038]
In addition, route search and the like can be performed by an external center. The center searches for the optimum route from the current position and destination of the vehicle, and supplies data necessary for route guidance to the vehicle side. This greatly simplifies the device on the vehicle side. In addition, when a roadside beacon is used for communication, transmission of the present location is unnecessary.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an overall configuration.
FIG. 2 is a flowchart illustrating an operation.
[Explanation of symbols]
10 navigation control unit, 12 own vehicle position detection unit, 14 map database, 16 speech synthesis unit, 18 input operation unit, 20a, 20b camera, 22a, 22b image processing unit, 24 recognition unit, 26 position relation measurement unit.

Claims (6)

In traveling with a map database that stores map information including position information of branch points and current position detection means for detecting the position of the host vehicle that is the current position of the host vehicle and a route to the destination is set A voice guidance device for a vehicle that outputs voice guidance for route guidance at a guidance branch point that is a branch point on the route and requires guidance,
Based on the detection result of the current position detection means and the map information stored in the map database, guidance branch point recognition means for recognizing that the guide branch point has been approached,
An in-vehicle camera that images the front,
A feature extraction means for extracting a feature used to identify the guidance branch point in voice guidance from an imaging result of the in-vehicle camera before the guidance branch point;
Positional relationship measuring means for measuring the positional relationship between the extracted feature and the own vehicle;
The positional relationship between the position of the guidance branch point read from the map database and the vehicle position detected by the current position detection unit, and the feature and the vehicle position measured by the positional relationship measurement unit Determination means for determining that the feature is a feature existing in the vicinity of the guide branch point based on the positional relationship between the guidance branch point and the feature obtained from both of the positional relationship ;
With
A voice guidance in which a guidance branch point is specified based on information on a feature determined to be present near the guidance branch point by the determination unit is included in the voice guidance of the route guidance at the guidance branch point and is output. A voice guidance device for a vehicle using image information. The apparatus of claim 1 .
The vehicle voice guidance apparatus using image information, wherein the positional relationship measuring means measures the distance between the vehicle and the feature detected by stereo ranging of a plurality of in-vehicle cameras. The apparatus according to claim 1 or 2 ,
The vehicle voice guidance apparatus using image information, wherein the feature is a traffic light. The apparatus of claim 3 .
An audio guidance device for a vehicle using image information, wherein the traffic light extracts a specific color portion and then specifies and detects the shape of the specific color portion by edge detection. The apparatus according to claim 3 or 4 ,
A voice guidance device for a vehicle using image information, wherein color information in a traffic light is added to the voice guidance. The apparatus according to claim 1 or 2 ,
The vehicle voice guidance apparatus using image information, wherein the feature is a destination guidance sign.

Images